RESUMO
BACKGROUND AND AIM: There are conflicting data on adiponectin concentrations in nonalcoholic fatty liver disease (NAFLD). The main aim was to compare circulating total adiponectin concentrations in NAFLD patients with versus without liver fibrosis. METHODS: A systematic search was performed in PubMed, Scopus, and Cochrane Library. Twenty-two studies comprising 1753 biopsy-proven NAFLD individuals (1290 with and 463 without fibrosis) were included in the meta-analysis. RESULTS: There was no difference in adiponectin concentration between NAFLD patients with versus without fibrosis (standardized mean difference [SMD]: -0.15; 95% confidence interval [95% CI]: -0.35 to 0.05). Heterogeneity was moderate among studies (Ι2 : 60%, P < 0.001); no risk of publication bias was observed (Egger's test; P = 0.37). The sensitivity analysis, performed after the exclusion of studies with (i) children/adolescents and morbidly obese patients (n = 3) and (ii) adiponectin measurement with other methods than enzyme-linked immunosorbent assay (ELISA) (n = 9), revealed significantly lower adiponectin concentrations in NAFLD patients with fibrosis (i) SMD: -0.23, 95% CI: -0.41 to -0.04; (ii) SMD: -0.30, 95% CI: -0.55 to -0.04, respectively). Meta-regression analysis revealed no significant association of adiponectin SMD with age, alanine aminotransferase, aspartate aminotransferase, γ-glutamyl-transferase, homeostasis model assessment insulin resistance and the proportion of men. CONCLUSIONS: Overall, patients with NAFLD and fibrosis had similar adiponectin concentrations with patients with NAFLD without fibrosis. However, adiponectin concentration was lower in NAFLD patients with fibrosis than those without fibrosis within the adult patients without morbid obesity and in studies in which adiponectin was measured with the same method (ELISA).